The proposed study will evaluate the relation between several current measures and indicators of alcoholism prevalence in a rural population for the purpose of providing statistically more accurate methods of planning preventive interventions. In this comparative statistical study, the results of an already completed interview study will be analyzed to derive behavioral indexes of drinking patterns. The study sample, composed of 2,998 randomly selected persons between the ages of 18 years and 90 years in Washington County, Maryland, responded to questions concerning their drinking habits, problems with drinking, and consequences of their drinking behavior. The operationally defined Quantity-Frequency index of heavy drinking will be used to epidemiologically compare these respondents' replies with the Mulford-Miller Index of Uncontrolled Drinking, the Cahalan, Cisin, and Crossley Index of Troubles Due to Drinking, and other behavioral prevalence estimators. The behavioral prevalence estimates will be compared with theoretical estimates and a predictive model will be developed; interaction effects of the various independent variables will also be identified in the course of the analysis. It is assumed that this study will allow for a meaningful comparison of the several prevalence indicators used widely in the literature as well as in the field for the purpose of planning and policy formulation. This study will, then, evaluate the reliability and validity of these indicators.